►
From YouTube: 2023-01-26 meeting
Description
cncf-opentelemetry meeting-2's Personal Meeting Room
A
Not
a
very
exciting
proposal
just
to
put
some
text
decimal
text
digits
together,
but
my
assignment
at
work
right
now
is
completely
in
different
place.
So
I'm
I'm
here
to
say
hi
but
I,
don't
think
I
had
much
to
say
about
sampling.
B
Foreign
yeah
I
also
have
nothing
on
the
agenda,
but
if
I
remember
correctly,
we
were
discussing
what
is
the
best
way
to
encode
the
adjust
account
right
for
yeah
Jerry
sampling
rates,
right
yeah.
A
B
And
you
proposed
I
think
some
kind
of
encoding
which
allows
for
some
sampling
rates
a
more
compact
representation.
A
Halfway
through
presenting
it,
I
kind
of
felt
like
that
was
maybe
a
step
too
far.
Just
having
the
leading
digits
of
the
significant
essentially
and
the
dropping
the
trailing
digits
would
be
good
enough.
I
I
had
in
my
sort
of
hackathon
projects,
just
put
the
number
of
leading
zeros
into
a
separate
counter,
essentially
to
to
encode
it,
but
I
I
think
that's,
probably
not
as
valuable
as
I
had
it
in
my
head
at
the
moment.
So
you
know
leading
zeros
could
be
included
and
then
any
significant
figures
after
that.
B
Sorry,
I
think
it
would
be
good
if
you
could
could
write
that
somehow
down
or
then
we,
it's
easier,
I,
think
to
understand
what
you,
what
you're
proposing,
because.
A
B
Enough,
probably
I
didn't
got
it
correctly
or
completely
and
all
right,
but
that's.
A
A
Piece
of
feedback
morning,
taliana
I
I,
was
just
mentioning
a
sort
of
hackathon
project.
I
did
and
pitched
two
weeks
ago,
and
it
it's
probably
worth
me
following
through,
but
I
haven't
so
at
the
moment.
I've
got
not
much
more
to
talk
about
there.
He
Ottmar
just
mentioned
that
he
would
prefer
to
see
a
nicer
write-up,
which
is
a
fair
piece
of
feedback,
so
I'll
I'll
shelve
that
it
was.
A
It
was
a
technique
proposing
to
use
the
supposed
randomness
of
the
w3c
Chase
parent
header
and
it
and
it's
in
the
agenda
from
last
time.
If
you
care
to
look
through
the
notes
and
click
on
it,
there's
just
a
GitHub
PR,
with
with
a
proof
of
concept
and
I
would
call
it
fairly
straightforward.
A
A
Now
that
we
have
kalyana
I
know
there
was
another
topic
that
was
kind
of
shelved
two
two
minutes
ago,
that
might
work
be
worth
a
refresher
or
otherwise.
I
have
no
agenda
items
here.
The
topic
was
instruments.
The
messaging
Sig
had
been
discussing,
span,
links
and
sampling
I,
don't
believe
that
we
got
much
traction
from
them.
A
There
was,
it
seemed
like
I
was
trying
to
sell
a
really
hard
idea.
That's
that's
my
summary.
So
the
idea
of
considering
the
wink
structure
remains
I,
guess
a
area
for
future
work
again.
There's
a
paper
trail
of
ideas,
but
it
doesn't
seem
like
people
are
very
excited
about
it,
or
at
least
the
problems
facing
that
group
are
bigger
more
important
than
the
idea
of
trying
to
get
your
spam
link
to
to
contribute
to
the
sampling
decision.
A
My
my
company
has
had
a
hand
in
that
Crush
in
that
discussion
outside
of
me,
but
the
the
reason
why
they're
interested
in
it,
though,
is
not
the
nice
thinking
about
sampling,
it's
more
like
the
Lambda,
the
Amazon
lambdas,
and
this
sort
of
serverless
function.
Execution
engines
out
of
the
world
are
difficult
to
get
traced
properly,
especially
when
messaging
is
taking
place
and
I.
Think
the
bigger
problem
that
they're
trying
to
solve
is
somewhat
bigger
than
the
idea
of
sampling.
Essentially
so
I
don't
have
much
of
a
report.
A
I
realize
it's
been
three
weeks
since
that
happened,
and
it's
another
group
at
my
company
who's
concerned
about
it.
So
I
have
nothing.
A
I
love
this
meeting
and
I
like
to
keep
it
going,
but
sometimes
there's
just
not
much
to
talk
about
I
I.
Keep
telling
my
management
that
my
most
exciting
opportunity
in
hotel
would
be
to
work
on
sampling.
After
my
current
assignment
so
right
now,
I'm
focused
on
the
Apache
Arrow
integration
that
we're
proposing
so
totally
different
area
and
well
nice
to
see
you
all
I
guess
we
could
probably
call
the
meeting
unless,
unless
anyone
has
a
topic.
B
No
thanks
I
just
want
to
lose
yeah,
let's,
let's
I.
A
C
You
out
last
week,
I
was
out,
I
mean
last
two
weeks.
I
was
out
sick
and
I
need
to
catch
up
on
all
the
recent
updates,
I
think
mainly
the
the
messaging
proposals
that
you
had
with
the
when
we
had
it
in
December
sometime.
The
meeting
where
so
I
will
I
will
go
catch
up
on
it
before
yeah.
A
C
One
small
question
I
want
to
ask,
is
I,
have
a
few
small
like
like
a
proof
of
Concepts
or
more,
like
example,
examples
that
I
built
mainly
for
say
tail
based
sampling,
or
even
this
link
link
based
sampling
things
like
that
or
even
stratified
sampling.
C
C
A
That
sounds
good.
You
know
I
when
I
when
I
was
writing
the
go
prototype
for
the
probability
sampling
that
we
have
I
put
it
in
the
Goku
trib
repository
I
think
at
Mars
also
merged
some
stuff,
so
yeah
that
sounds
good.
If
there's
a
net
contrib,
you
could
put
it
there.
I
would
say
the
bar
is
going
to
be
low.
Enough
depends
on
the
maintainers,
yeah
and
I
would
love
to
to
be
a
reviewer
on
that
as
I.
Imagine,
anyone
in
this
room
would
be.